Songhao Wu
shwu
AI & ML interests
Mixture-of-Experts Model, Language Model Pretraining
Recent Activity
upvoted a paper about 15 hours ago
Periodic Weak Spots: Phase Sensitivity from Chunked KV-Cache Compression liked a model 9 days ago
IQuestLab/IQuest-Q1 upvoted a paper 4 months ago
LatentMoE: Toward Optimal Accuracy per FLOP and Parameter in Mixture of ExpertsOrganizations
None yet